Retraction: A growing, if uncomfortable, state of affairs

Retraction of peer review published articles has become an increasing activity not only in the sciences, but in both social sciences and in management. They are up by an order of magnitude in recent times. A global flood of faculty are under increasing pressure to publish in top journals whose publication ‘slots’ haven’t increased in accordance with demand.

One result, it would seem, are attempts to ‘cut corners’, in order to enhance the publication prospects of our work. The consequences may lead to erroneous publication, however, the sword of Damocles hangs over all our heads, leading to retraction, as well as significant public and professional embarrassment.

We have recently observed a number of well known cases in the field of management, including both a German Professor and an American scholar who had multiple publications retracted from leading management journals. Both cases engendered considerable controversy, rebuke, counter rebuke, litigation and professional frustration, as co-authors, senior and jr. scholars, editors, and reviewers found themselves involved in heated debates regarding the merits, ethics, and legal rights surrounding these retractions.

This blog cannot adjudicate regarding the decisions made by editors or publishers that eventually lead to retraction, nor is that our goal. However, it is worth noting that our own code of ethics addresses this issue squarely; by stating:

“if AOM members discover significant errors in their publication or presentation of data, they take appropriate steps to correct such errors in the form of a correction, retraction, published erratum, or other public statement”.

The site, retraction watch is a fountain of information regarding when, how and why these actions take place. A MacArtthur foundation grant insures that the founding writers, Ivan Oransky and Adam Marcus, will be reporting on the ethical errors, omissions, and sometimes outrageous antics of our profession for some time to come. They have called for a journal transparency index asking journals to formally codify many of the ethical guidelines that already exist in our own Code of Ethics

They include:

  • The journal’s review protocol, including whether or not its articles are peer-reviewed; the typical number of reviewers, time for review, manuscript acceptance rate, and details of the appeals process
  • Whether the journal requires that underlying data are made available
  • Whether the journal uses plagiarism detection software and reviews figures for evidence of image manipulation
  • The journal’s mechanism for dealing with allegations of errors or misconduct, including whether it investigates such allegations from anonymous whistleblowers
  • Whether corrections and retraction notices are as clear as possible, conforming to accepted publishing ethics guidelines such as those from the Committee on Publication Ethics (COPE) or the International Committee of Medical Journal Editors (ICMJE)

As management scholars, surprisingly, we are only beginning to learn the importance of some of these critical issues related to the scientific method including replication, data recording and data sharing, enabling others to verify the accuracy of our analyses.

I personally recall a senior editor writing that because replication was of no interest to their journal, a replicated and extended finding that demonstrated one of their previously well cited publications was erroneous was not within the journal’s mandate, of no concern, and thus subject to immediate desk rejection. Taking the point of view of this editor, one might determine that retractions or replications that undermine previously published work only serve to detract from a journal’s reputation.   However, from a field perspective, our contribution to science is predicated on our ability to adopt and ensure that rigorous standards are adhered to, even of the cost seems unreasonable to the average scholar (such as a decision to retract erroneous work).

No doubt, there will be an increasing number of ‘incidents’ of retraction as our scholarly community begins to embrace something akin to a transparency index, and as we more literally interpret AOM’s code of ethics regarding research and publication. While the process may be somewhat painful, I look forward to the outcome in terms of increased scientific reliability.

2 thoughts on “Retraction: A growing, if uncomfortable, state of affairs”

  1. This is a very important topic. I think it’s vital, however, to distinguish the problem of retraction from the issue of replication. A published result does not have to be retracted merely because a replication fails. We do have to be much more open to the publication of failed replications, but not–and this is my point–because they should get us thinking about retracting the original study (the one that could not be replicated). When a result cannot be replicated, that does not automatically suggest ethical (or even methodological) problems in the original study. It’s just (or ought to be) a normal part of science.

    To be sure, a failed replication might in some cases indicate that something more fishy was going on in the original study. But until further inquiries into the conditions under which the original data was generated, or the means by which the original analysis was done, suggests problems, there’s no need to think about retracting it. (There are of course also cases in which deeper problems can be discerned, as it were, “on the surface” of the published results. Uri Simonsohn is the Zen master of this sort of critique.) In fact, since retracting is often socially embarrassing and sometimes legally complicated, I think it’s important not to associate it too directly with replication. It might make too many researchers shy away from replicating the work of others, for fear of getting into thorny, sensitive territory.

    The dearth of replications is arguably a more serious problem for the managerial sciences than the surge of retractions.

  2. I agree whole-heartedly. There are many reasons why a replication may fail, in most cases, having no relationship to retraction whatsoever. However, on a few occasions, replications have alerted scholars to fraud, eventually leading to retraction. Most importantly, making errors is a normal part of any research. Time and the advance of methods and studies provide additional perspectives. We must move beyond the view that our published research is sacrosanct. This means editors must take a broader view regarding what is relevant to their readership, irrespective of concerns for their journal’s ‘impact factor’. As our code points out, there are many ways to acknowledge errors in research, and retraction is just one. Only when we acknowledge our errors, however, can we improve our science.

Leave a Reply